Date: 2024-09-11


Use the same set of confounders as in the main analysis

confounders_infection <- c(
  "age_at_test_in_months_cat_2", # 3-month interval
  "month_when_tested_cat"
)

confounders_ed <- c(
  "age_at_test_in_months_cat_2", # 3-month interval
  "month_when_tested_cat"
)

confounders_inpatient <- c(
  "age_at_test_in_months_cat_2", # 3-month interval
  "month_when_tested_cat",
  "risk_factor_atleastone"
)

confounders_severe <- c(
  "age_at_test_in_months_cat_2", # 3-month interval
  "month_when_tested_cat",
  "risk_factor_atleastone"
)

confounders_LRTI <- c(
  "age_at_test_in_months_cat_2", # 3-month interval
  "month_when_tested_cat",
  "risk_factor_atleastone"
)

confounders_LRTIhosp <- c(
  "age_at_test_in_months_cat_2", # 3-month interval
  "month_when_tested_cat",
  "risk_factor_atleastone"
)


VE over time by biweek interval with imposed monotonic trend

  • classify time since vax to testing by biweek interval.
  • included confounders : age tested, month tested
  • uninformative prior for precision of the non-negative increment d’s.
    • d ~ dnorm(0, tau)T(0,)
    • tau_d ~ dgamma(0.01, 0.01)

Table

Visualize the results

One endpoint per panel

Overlap VE waning curve from Hodgson et al. 2024 with VE here against corresponding endpoint

traceplot